Welding inverter barmaley. Welding inverter barmaley Simple single-stroke welding inverter oblique bridge

Recently I assembled a welding inverter from Barmaley, for a maximum current of 160 amperes, a single-board version. This scheme is named after its author - Barmaley. Here is the wiring diagram and PCB file.

Inverter circuit for welding

Inverter operation: power from a single-phase 220 Volt network is rectified, smoothed by capacitors and fed to transistor switches, which make a high-frequency variable from a constant voltage, supplied to a ferrite transformer. Due to the high frequency, we have a reduction in the size of the power trance and, as a result, we use not iron, but ferrite. Next is a step-down transformer, followed by a rectifier and a choke.

Oscillograms control field-effect transistors. Measured on a zener diode x213b without power switches, filling factor 43 and frequency 33.

In its version, the power keys IRG4PC50U replaced with more modern IRGP4063DPBF... The ks213b zener diode was replaced by two 15-volt 1.3 watt counter-connected ones, since in the previous ks213b device warmed up a little. After the replacement, the problem disappeared immediately. The rest remains as in the scheme.

This is an oscillogram of the collector-emitter of the lower key (according to the diagram). When power is supplied to 310 volts through a 150 watt lamp. The oscilloscope costs 5 volt division and 5 μs div. through the divisor multiplied by 10.

The power transformer is wound on the core B66371-G-X187, N87, E70 / 33/32 EPCOS Winding data: first the floor of the primary, the secondary, and again the remnants of the primary. The wire on the primary and secondary is 0.6 mm in diameter. Primary - 10 wires 0.6 twisted together 18 turns (total). 9 turns just fit into the first row. Further, the remains of the primary to the side, we wind 6 turns with a 0.6 wire folded into 50 pieces, also twisted. And then again the remains of the primary, that is, 9 turns. Do not forget the interlayer insulation (used several layers of cash paper, 5 or 6, we are no longer zealous, otherwise the winding will not fit into the window). Each layer was impregnated with epoxy.

Then we collect everything, a gap of 0.1 mm is needed between the halves of the E70 ferrite, on the extreme cores we put a gasket from a regular cashier's receipt. We pull everything together, glue it together.

I spray-painted with matte black paint, then varnish. Yes, I almost forgot, when we twisted each winding, we wrap it with paint tape - we insulate it, so to speak. Do not forget to mark the beginning and ends of the windings, it will be useful for further phasing and assembly. If the transformer is phased incorrectly, the apparatus will cook at half strength.

When the inverter is connected to the grid, the charging of the output capacitors begins. The initial charging current is very high, comparable to a short circuit, and can lead to burnout of the diode bridge. Not to mention the fact that for the conductors it is also fraught with failure. To avoid such a sharp jump in current at the moment of switching on, capacitor charge limiters are installed. In Barmaley's circuit, these are 2 resistors of 30 ohms, with a power of 5 watts, totaling 15 ohms x 10 watts. The resistor limits the charging current of the capacitors, and after charging them, you can already supply power directly, bypassing these resistors, which is what the relay does.

The WJ115-1A-12VDC-S relay is used in the Barmaley welding machine. Power supply of the relay coil - 12 volts DC, switching load 20 amperes, 220 volts AC. In homemade products, the use of automotive relays for 12 Volts, 30 Amperes is very common. However, they are not intended for switching currents up to 20 Amperes of mains voltage, but, nevertheless, they are cheap, available and do their job quite well.

It is better to install a current-limiting resistor with a conventional wire-wound resistor, it will withstand any overloads and is cheaper than imported ones. For example C5-37 V 10 (20 Ohm, 10 Watt, wire). Instead of resistors, you can put current-limiting capacitors in series in the AC voltage circuit. For example K73-17, 400 Volts, with a total capacity of 5-10 μF. Capacitors 3 uF, charge a capacity of 2000 uF, in about 5 seconds. The calculation of the capacitor charging current is as follows: 1 μF limits the current to 70 milliamperes. It turns out 3 uF at the level of 70x3 \u003d 210 milliamperes.

Finally, I put everything together and launched it. The limited current set 165 amperes, now we will arrange the welding inverter in a good case. The cost of a home-made inverter is about 2,500 rubles - I ordered the parts on the Internet.

I took the wire in the rewinding shop. You can also remove the wire from the TVs from the demagnetizing circuit from the kinescope (this is almost a finished secondary). The choke is made from E65, copper strip 5 mm wide and 2 mm thick - 18 turns. I picked up the inductance 84 μH by increasing the gap between the halves, it was 4 mm. You can not wind it with a strip, but also with 0.6 mm wire, but it will be more difficult to lay it. The primary on the transformer can be wound with a 1.2 mm wire, a set of 5 pieces of 18 turns, but you can also calculate 0.4 mm the number of wires for the section you need, that is, for example, 15 pieces of 0.4 mm 18 turns.

After installing and adjusting the circuit on the board, I put everything together. Barmaley's tests were successful: he pulls the electrode three and four calmly. The limiting current delivered 165 Amperes. Collected and tested the device: Arsi .

Discuss the article BARMALE WELDING INVERTER

The welding inverter is a fairly popular device that is necessary both in the household and in an industrial enterprise. This is not surprising, because the power supplies that were used before (converters, transformers, rectifiers) had many disadvantages. Among them are weight and dimensions, high energy consumption, but a small range of welding mode control and low conversion frequency. Having made a thyristor-based welding inverter with your own hands, you will receive a powerful power supply for the necessary work. It will also help you significantly save money, although it will still require certain labor and material costs.

Welding inverter: features and functions of the device

The job of an inverter is to convert the AC mains current to its DC high frequency counterpart.

This takes place in several stages. Current flows to the rectifier unit from the network. There, after transformation, the voltage from alternating becomes constant. And the inverter performs the reverse conversion, that is, the incoming DC voltage again becomes alternating, but with a higher frequency. After that, the voltage is reduced by a transformer, through the output rectifier, this parameter is modified into a high-frequency constant voltage.

The design of the welding inverter and its features

Due to the fact that there are no heavy parts in the construction of the device, it is very compact and lightweight. It includes the following components:

Simple cross-coupled inverter device.

• inverter;
• mains and output rectifiers;
• throttle;
• high frequency transformer.

Even novice welders can work with such machines. They are used both in everyday life and in the construction industry or in car services. Due to the fact that there is an adjustment of operating modes, both thin and thick metals can be cooked. And the increased conditions of arc burning and formation of a weld seam give you the opportunity to weld any alloys, ferrous and non-ferrous metals with welding inverters, using all possible welding technologies.

Benefits of using an inverter

In the field of welded equipment, such devices are in special demand due to their many advantages and benefits. Having made an inverter with your own hands, you will get:

• the ability to cook complex non-ferrous metals and structural steels;
• protection against overheating, mains voltage fluctuations, current overloads;
• high stability of the welding current, even though the voltage may fluctuate in the network;
• well-formed seam;
• there will be practically no spatter when welding;
• arc burning will be stabilized in a given way, even if there is an external adverse effect;
• many other useful functions.

Diy inverter circuits

Taking as a basis how the circuit is built and how the inverter conversion process itself is controlled, several types of devices are distinguished, which are the most common in use. The full bridge and half bridge options are referred to as two push-pull circuits, while the oblique bridge is referred to as single-cycle. The full bridge circuit, called push-pull, operates with bipolar pulses. They are fed to the key transistors (which are paired), and they turn off and open the electrical circuit.

Inverter circuit "oblique" bridge.

The half-bridge circuit will differ from the previous version in that its current consumption is increased. Transistors operating according to the same push-pull model act as keys. Each of them is supplied with half of the mains input voltage. The power of the inverter, compared to the current with a full bridge, is half the value. This arrangement has its advantages in low power applications. In addition, you can use a group of transistors, and not one very powerful one.

The last option is the "oblique" bridge. These are inverters that operate on a one-cycle basis. Here you will be dealing with unipolar impulses. Simultaneous opening of transistor switches will eliminate the possibility of a short circuit. But among the disadvantages of this scheme, the bias of the magnetic circuit of the transformer is distinguished.

Take a look at one of the standard inverter circuits. This is a structure designed by Yu. Negulyaev. To assemble such a device at home, you will need your desire, readiness for work and the necessary element base, which you can either find on the radio market, or evaporate from old household appliances.

Assembly instructions

Standard inverter circuit designed by Yu.Negulyaev

Take a 6mm duralumin slab. Attach all conductors and wires that give off heat to it. Please note that here the wire does not need to be wrapped with heat-insulating material. Using an old circuit (for example, a computer), you do not have to separately search for transistors and thyristors.

Next, prepare a special high-power fan (you can even use a car radiator). It will blow everything, including the resonant choke. Remember to press the latter against your base with a gasket seal.

For the manufacture of the choke device itself, take six copper cores. They can be found on the market or made yourself from parts of an unnecessary old TV. Press the diodes to the base of the circuit, and then attach the voltage regulators and isolation seals to them.

When installing the transformer, insulate the conductor bundles with electrical tape or fluoroplastic tape. Separate the conductors in different directions so that they do not contact and do not cause malfunctions. On the field-effect transistor, you will need to install a force field in order to extend the performance of your inverter. To do this, take a 2 mm copper wire. After tinning it, wrap it in several layers with ordinary thread. This will protect your conductor from various types of damage during soldering and welding. Use insulating heels to secure the mounting. So you will also transfer the load from the transistors to them.

Power circuit with power supply and drivers.

………. The welding inverter shown in the diagram is built according to the scheme of a single-stroke forward feed. Unipolar pulses of rectified mains voltage with a filling of not more than 42% are supplied to the primary winding of a welding transformer with the help of two keys. The magnetic circuit of the transformer undergoes one-way bias. In the pauses between pulses, the magnetic circuit is demagnetized along the so-called private loop. The demagnetizing current, thanks to the diodes switched back on, returns the magnetic energy stored in the transformer core back to the source, recharging the capacitors (2 x 1000 μF x 400 V) of the drive.

………. On the forward run, energy is transferred to the load through the welding transformer and directly connected rectifier diodes (2x150EBU04). During the pause between pulses, the current in the load is maintained by the energy stored in the inductor. The electrical circuit in this case is closed through the check diodes (2x150EBU04). It is well known that these diodes have a greater load than direct ones, because the current flows longer in a pause than in a pulse.

………. A 1200 μF x 250 V capacitor connected to the welding wires through a 4.3 Ohm resistor ensures a clear ignition of the arc. Perhaps this is one of the most successful circuit solutions for ignition in a space bridge.

………. Oblique bridge keys operate in hard switch mode. Moreover, the switching mode is deliberately facilitated by the always present leakage inductance of the welding transformer. And, since by the time the keys are turned on, it is considered that the magnetic circuit of the transformer is completely demagnetized, due to the lack of current in the primary winding, turn-on losses can be neglected. Loss of shutdown is very significant. To reduce them, RCD snubbers are installed in parallel to each key.

………. To ensure the precise operation of the keys, a negative voltage is applied to their gates between the switches on due to a special circuit for switching on the drivers. Each driver is powered by a galvanically isolated source (about 25 V) from the power supply. The supply voltage of the "upper" driver is used to turn on relay K1, whose contacts bypass the starting resistor.

………. The power supply (classic low-power flyback) has 3 galvanically isolated outputs. With intact parts, it starts working immediately. The voltage for the drivers is 23-25V. 12 V is used to power the control unit.

………. Substantial heat sinks must be provided for the input rectifier, switches, and output rectifier. The operating time of the device will depend on the size of these radiators and the intensity of their blowing. Since the device provides a substantial welding current (up to 180 A), the keys must be soldered to copper plates 4 mm thick, then these "sandwiches" must be screwed to the radiators through heat-conducting paste. How to do it is written Together for fastening the keys, the radiator seat should be ideally flat without chips and shells. It is desirable that the radiator has a solid body at least 10 mm thick at the place where the keys are attached. As practice has shown, for better heat dissipation, it is not necessary to isolate the keys from the radiator. Better to isolate the radiator from the body of the device. The blower must also be supplied with a transformer, a choke and necessarily all resistors with a power of 25 and 30 W. The rest of the circuit elements do not need radiators and airflow.

Control block

Diagram of the control unit for full-bridge welding inverter

………. The control unit is built on the basis of the popular TL494 PWM controller with one control channel. This channel stabilizes the current in the arc. The current reference drives the microcontroller using the CCP1 module in PWM mode at a frequency of approximately 75 kHz. The PWM filling will determine the voltage at C1. The magnitude of this voltage determines the magnitude of the welding current.

………. The microcontroller also blocks the inverter. If the DT (4) input of the TL494 is set to a high logic level, the pulses on Out will disappear and the inverter will stop. The appearance of a logical zero at the RA4 output of the microcontroller will lead to a smooth start of the inverter, that is, to a gradual increase in the filling of pulses at the Out output to the maximum. The blocking of the inverter is used at the moment of switching on and when the temperature of the radiators is exceeded.

Here's what happened in hardware. Power supply, drivers and control unit on one board.


. In my device, the indicator and keyboard are connected to the control unit via a computer loop. The loop runs in the immediate vicinity of the key radiators and the transformer. In its pure form, such a construct led to false keystrokes. I had to apply the following specials. measures. Ferrite ring K28x16x9 on the train. The train is twisted (as far as its length allowed). For the keyboard and thermostats, additional 1.8K pull-up resistors are used, shunted with 100 pkf ceramic capacitors. Such a circuit solution provided the keyboard noise immunity, and false keystrokes were completely excluded.

………. Although, in my opinion - you need to avoid interference in the control unit. For this, the control unit must be separated from the power unit by a solid sheet of metal.

Inverter setting

………. The power section is still de-energized. We connect the previously tested power supply to the control unit and plug it into the network. All the eights will light up on the indicator, then the relay will turn on and, if the thermostat contacts are closed, the indicator will show the current setting of 20 A. With the oscilloscope, we check the voltage at the keys. There should be rectangular pulses with fronts of no more than 200 ns, with a frequency of 40-50 kHz and a voltage of 13-15 V in the positive region and 10 V in the negative. Moreover, in the negative region, the pulse should be noticeably longer.

………. If everything is so, we assemble the entire inverter circuit and plug it into the network. At first, the display will display eights, then the relay should turn on and the indicator will show 20 A. By clicking the buttons, we try to change the current setting. Changing the current reference must proportionally change the voltage across the capacitor C1. If, after changing the current reference, the buttons are not pressed for more than 1 minute, the reference will be written to the non-volatile memory. The indicator will briefly show the message “REGISTER”. The next time the inverter is turned on, the current reference value will be equal to the value that was written.

………. If everything is so, we set the task to 20 A and turn on a load rheostat with a resistance of 0.5 Ohm into the welding wires. The rheostat must withstand a current flow of at least 60 A. change the current setting, and according to the voltmeter readings, we control the current. In this mode, the rheostat can emit a ringing sound. They should not be afraid of him - this is the current limitation. The current should vary in proportion to the task. We set the current setting to 50 A. If the voltmeter readings do not correspond to 50 A, then on the switched off inverter we solder the resistance R1 of a different rating. By selecting the resistance R1, we achieve the correspondence of the current setting to the measured one.

………. We check the work of thermal protection. For this we break off the thermostat circuit. The indicator will show the inscription "EroC". The pulses on the key gates should disappear Restoring the thermostat circuit. The indicator should show the set current. Pulses should appear on the key locks. Their duration should gradually increase to maximum.

………. If everything is so, you can try to cook. After 2-3 minutes of welding with a current of 120-150 A, turn off the inverter from the network and look for the 2 hottest radiators. Protective thermostats must be installed on them. If possible, thermostats are installed outside the blowing zone.

Quite often, the main three types of high-frequency converters are used to build a welding inverter, namely, converters connected according to the schemes: asymmetric or oblique bridge, half bridge, and full bridge. In this case, resonant converters are subspecies of half-bridge and full-bridge circuits. According to the control system, these devices can be divided into: PWM (pulse width modulation), PFM (frequency control), phase control, and combinations of all three systems can also exist.

All of the above converters have their pros and cons. Let's deal with each separately.

PWM half-bridge system

The block diagram is shown below:

This is, perhaps, one of the simplest, but no less reliable converters of the push-pull family. The "swing" voltage of the primary winding of the power transformer will be equal to half the supply voltage - this is a drawback of this circuit. But if you look from the other side, you can use a transformer with a smaller core, without fear of entering the saturation zone, which is also a plus. For welding inverters with a power of about 2-3 kW, such a power module is quite promising.

Since power transistors operate in hard switching mode, drivers must be installed for their normal operation. This is due to the fact that when operating in this mode, the transistors need a high quality control signal. It is also necessary to have a currentless pause in order to prevent the simultaneous opening of transistors, which will result in the failure of the latter.

A rather perspective view of a half-bridge converter, its circuit is shown below:

A resonant half bridge will be slightly simpler than a PWM half bridge. This is due to the presence of a resonant inductance, which limits the maximum current of transistors, and the switching of transistors occurs at zero current or voltage. The current flowing through the power circuit will be sinusoidal, which will take the load off the capacitor filters. With this circuit design, drivers are not necessary, switching can be carried out with a conventional pulse transformer. The quality of the control pulses in this circuit is not as important as in the previous one, but there should still be a currentless pause.

In this case, you can do without current protection, and the form of the current-voltage characteristic, which does not require its parametric formation.

The output current will be limited only by the magnetizing inductance of the transformer and, accordingly, will be able to reach quite significant values, in the event that a short circuit occurs. This property has a positive effect on the ignition and burning of the arc, but it must also be taken into account when selecting the output diodes.

Typically, the output parameters are controlled by changing the frequency. But phase control also gives some of its advantages and is more promising for welding inverters. It allows you to bypass such an unpleasant phenomenon as the coincidence of the short circuit mode with resonance, and also increases the control range of the output parameters. The use of phase control can allow you to change the output current in the range from 0 to I max.

Asymmetric or "oblique" bridge

This is a single-ended, forward converter, the block diagram of which is shown below:

This type of converter is quite popular with both ordinary radio amateurs and manufacturers of welding inverters. The very first welding inverters were built exactly according to such schemes - an asymmetric or "oblique" bridge. Noise immunity, a fairly wide range of output current regulation, reliability and simplicity - all these qualities still attract manufacturers to this day.

Quite high currents passing through transistors, an increased requirement for the quality of the control pulse, which leads to the need to use powerful drivers to control transistors, and high requirements for installation work in these devices and the presence of large pulse currents, which in turn increase the requirements for - these are significant disadvantages of this type of converter. Also, to maintain the normal operation of the transistors, it is necessary to add RCD chains - snubbers.

But despite the above-listed disadvantages and low efficiency of the device according to the asymmetric or "oblique" bridge scheme, they are still used in welding inverters. In this case, transistors T1 and T2 will work in phase, that is, they will close and open at the same time. In this case, the accumulation of energy will not occur in the transformer, but in the choke coil Dr1. That is why in order to get the same power with a bridge converter, double the current through the transistors is required, since the duty cycle will not exceed 50%. We will consider this system in more detail in the following articles.

It is a classic push-pull converter, the block diagram of which is shown below:

This circuit allows you to get power 2 times more than when turning on the half-bridge type and 2 times more than when turning on the "oblique" bridge type, while the currents and, accordingly, losses in all three cases will be equal. This can be explained by the fact that the supply voltage will be equal to the "swing" voltage of the primary winding of the power transformer.

In order to obtain the same power with a half-bridge (swing voltage 0.5U supply), a current is required 2 times! less than for the half-bridge case. In a full bridge circuit with PWM, transistors will operate alternately - T1, T3 are on, and T2, T4 are off and, accordingly, vice versa when the polarity changes. Through track and control the values \u200b\u200bof the amplitude current flowing through this diagonal. There are two most commonly used ways to regulate it:

• Leave the cutoff voltage unchanged, but change only the control pulse length;
• Carry out changes in the cut-off voltage level according to the data from the current transformer while leaving the control pulse duration unchanged;

Both methods can allow changes in the output current to be carried out within fairly large limits. A full PWM bridge has the same disadvantages and requirements as a PWM half bridge. (See above).

It is the most promising high-frequency converter circuit for a welding inverter, the block diagram of which is shown below:

A resonant bridge is not much different from a full PWM bridge. The difference is that with a resonant connection, a resonant LC circuit is connected in series with the transformer winding. However, its appearance radically changes the process of pumping power. Losses will decrease, efficiency will increase, the load on the input electrolytes will decrease, and electromagnetic interference will decrease. In this case, drivers for power transistors should be used only if MOSFET transistors are used, which have a gate capacity of more than 5000 pF. IGBTs can only get by with a pulse transformer. More detailed descriptions of the circuits will be given in the following articles.

The output current can be controlled in two ways - frequency and phase. Both of these methods have been described in a resonant half-bridge (see above).

Full bridge with dispersion choke

Its circuit is practically no different from the resonant bridge or half-bridge circuit, only instead of the LC resonant circuit, a non-resonant LC circuit is connected in series with the transformer. The capacitance C, approximately C≈22μF x 63V, works as a balancing capacitor, and the inductive reactance of the inductor L as reactance, the value of which will change linearly depending on the frequency change. The converter is controlled by the frequency method. , as the voltage frequency increases, the inductance resistance will increase, which will reduce the current in the power transformer. Quite a simple and reliable way. Therefore, a fairly large number of industrial inverters are built according to this principle of limiting the output parameters.

The arc welder must provide a falling current-voltage characteristic in the load (arc). In bridge inverters, as a rule, the falling characteristic is provided by rather complex electronics with mandatory current feedback. From the point of view of ease of control, in my opinion, the resonant bridge is the most attractive. In it, the falling characteristic of the welding current source is provided by the parametric properties of the resonant circuit in the primary circuit of the inverter.

A feature of the inverter presented in this article is not only the use of a full resonant bridge, but also control of it using the PIC16F628-20I / P microcontroller.

Immediately, we note that the maximum welding current of the inverter depends on the setting. Its value is entirely determined by the width of the non-magnetic gap in the magnetic circuit of the resonant choke. For the power elements used in the inverter, subject to their thermal conditions, the welding current can reach 200 A.

The circuit diagram of the inverter is divided into two parts. On fig. 1the power section is shown, and on fig. 2- a diagram of a power supply unit with a control unit. The classic bridge welding inverter consists of a mains voltage rectifier with filter capacitors. A constant voltage of 300 V with the help of 4 switches is converted into an alternating voltage of a higher frequency, which is lowered with a welding transformer and then rectified.

Power section

In resonant converters in series with the primary winding of the welding transformer T1, a resonant choke L1 and a resonant capacitor C1-C10 are connected (see. fig. 1where the power circuits are highlighted in bold). The inductance of the series circuit consists of the inductance of the resonant choke L1 and the inductance of the primary winding of the transformer T1. The secondary winding T1 is loaded with a welding arc. If the capacitance C1-C10 and the inductance L1 are constant values, then the inductance of the primary winding T1 depends on the load resistance in the secondary winding, i.e. from the welding current. The maximum inductance of the primary winding T1 corresponds to the "idle" mode of the inverter, and the minimum - to the short circuit mode. The load resistance also determines the quality factor of the circuit. Thus, the resonant frequency of the circuit is minimum in the "no-load" mode (with the maximum inductance of the primary winding T1) and maximum in the short-circuit mode (with the minimum inductance of the primary winding T1). When the inverter is loaded with a welding arc, the resonant frequency of the loop depends on the current in the arc.

From all of the above, it is obvious that the frequency of the inverter when operating at maximum power in the arc must be lower than the natural frequency of the resonant circuit of the inverter in the short circuit mode and higher than it in the "no-load" mode. It is optimal that resonance occurs at the natural frequency of the circuit, at which the maximum power develops in the arc (f MAX. POWER). This is the main criterion for the correct setting of the inverter. If in this case increase the frequency of the inverter relative to f MAX. POWERFUL. , the current in the arc decreases due to the increase in the inductive resistance of the resonant choke L1. This is how the frequency control of the current in the welding arc is carried out.

Resonance in the inverter circuit in the event of a short circuit and incorrect setting of the inverter is possible at a frequency higher than f MAX. POWERFUL. ...

Note also that resonance is unacceptable in the short circuit mode for transistor switches of the inverter due to the occurrence of an overcurrent in the primary circuit. Since the short-circuit mode is a normal mode for the welding machine, it is necessary to prevent the inverter from operating at frequencies higher than f MAX. POWERFUL. in case of a short circuit in the welding circuit.

To do this, in this inverter, a microcontroller continuously monitors the fact of a short circuit in the welding wires using a special detector. In the event of a short circuit, the microcontroller automatically reduces the frequency of the inverter to the previously set value f MAX. POWERFUL. - at this frequency, resonance in the short circuit is impossible, which prevents excessive current flow in the primary circuit and, accordingly, through the switches.

In the power section (fig. 1)R13 is the starting resistor. It limits the charging current of the oxide capacitors C16, C17 when the device is turned on. The diode bridge VD14-VD21 is designed to rectify the mains voltage 220 V / 50 Hz, which is smoothed by capacitors C15-C17 and fed to the output bridge of the circuit, consisting of 4 keys on IGBT transistors VT1-VT4.

Suppressors VD3, VD9 and VD22 protect the keys from voltage surges. Resistors R5, R6 discharge the resonant capacitor when the inverter is turned off. Zener diodes VD1, VD2, VD4, VD5 do not allow the voltage at the gates of the switches to exceed 18 V. Resistors R1, R3, R7 and R9 limit the output current of the drivers at the moments of charge-discharge of the gate capacitors of the switches. Resistors R2, R4, R8, R10 provide reliable key closure when there is no power to the drivers.

Welding transformer T1 with transformation ratio 6 lowers the voltage and provides galvanic isolation of the output from the mains part of the inverter. The alternating voltage from the secondary winding of the welding transformer is rectified by diodes VD6, VD7 and goes through the welding wires to the electrode and the surfaces to be welded. R11C13 and R12C14 circuits are used to absorb the energy of the reverse voltage surges of the output rectifier. For stable burning of the arc at low currents, as well as to facilitate its ignition, a voltage doubler is provided, assembled on the elements C11, C12, VD10-VD13, C19, C20 and L2. Resistor R14 serves as a doubler load. The VD8 suppressor protects the diodes of the output rectifier from reverse voltage surges.

Power Supply

Built according to a flyback converter circuit based on a specialized microcircuit DA6 TNY264 according to a typical circuit (fig. 2)... It provides power for drivers, relays and microcontroller control unit. The power supply for the drivers of the upper keys is galvanically isolated from the 24 V relay supply channel and the supply channel of the lower drivers. A parametric stabilizer DA7 is used to power the DD1 microcontroller (5 V). The DA1-DA4 drivers of the HCPL3120 type are designed to control the keys VT1-VT4 and provide steep edges of the control pulses on the gates of these transistors.

The short-circuit detector is assembled on the elements R25, R27, R28, DA8, VD32, VD33, C38. When the voltage on the welding wires is below 9 V (short circuit), a high logic level appears at the RB4 input of the DD1 controller, and when the voltage is more than 9 V (no short circuit), a low logic level appears at the RB4 input.

The DD1 position uses a widespread microcontroller (MC) PIC16F628-20I / P in a DIP package.

Inverter operation

As soon as the power supply starts up, the microcontroller program starts working. After a delay of about 5 seconds, the buzzer will sound and the inverter will start working. As soon as the voltage in the welding wires exceeds 9 V, the MK will open the VT5 key, which will turn on the K1 relay, and the relay contacts will shunt the charging resistor R13. The buzzer will also be muted. At this point, the inverter is ready for operation. The frequency of the inverter will be determined by the position of the potentiometer R18. Moreover, the minimum frequency (aka f MAX. POWER) corresponds to the maximum welding current, and the maximum frequency corresponds to the minimum current. The frequency changes stepwise (discretely). Only 17 positions are used. When the potentiometer R18 is rotated, the frequency change is accompanied by a short sound signal from the buzzer. Thus, you can change the frequency of the welding current by the sound of the buzzer by the required number of positions.

In the event of a short circuit in the welding leads, the inverter automatically starts operating at f MAX. POWERFUL. , - Inverter operation in short-circuit mode is accompanied by a buzzer sound signal. If the short circuit lasts more than 1 s, then the operation of the inverter is blocked and after 3 s it resumes again. This is how the anti-sticking function of the electrode is implemented.

In the absence of a short circuit, a low logic level is applied to the RB4 input, and the inverter frequency is determined by the position of the potentiometer R18.

To protect the output keys from overheating, two thermostats TS1 and TS2 are used as sensors. If at least one of the thermostats is disconnected, the operation of the inverter is blocked. The buzzer emits an intermittent rapid sound signal until the radiator cools down, on which the triggered thermostat is installed.

Construction and detailsResonant choke L1 is wound on a magnetic core ETD59, material No. 87 from EPCOS and contains 12 turns of copper wire with a diameter of 2 mm in varnish insulation. The wire is wound with a mandatory gap between the turns. Thick thread can be used to provide clearance. To fix the winding, you need to coat the turns with epoxy glue. The halves of the magnetic core are joined with a non-magnetic gap of 1 ... 2 mm. A more accurate value of the non-magnetic gap is selected when tuning the resonant frequency. During operation of the inverter, the resonant choke magnetic circuit may become very hot. This is due to the saturation of the ferrite during operation at resonance. To ensure reliable fixation of the gap of the magnetic core, its halves must be pulled together with metal pins. In this case, it is necessary to ensure that the distance from the gap to the studs is at least 5 mm. Otherwise, the studs near the gap may melt. For the same reason, it is unacceptable to tighten the choke with a solid metal casing.

Transformer T1 is wound on a magnetic core E65, material No. 87 from EPCOS. First, the primary winding is wound in one row - 18 turns of copper wire with a diameter of 2 mm in varnish insulation. Windings II and III are wound over the primary winding. Each of them takes up half of the frame. Windings II and III contain 3 turns in four copper wires with a diameter of 2 mm. The halves of the transformer magnetic circuit are joined without a gap and securely fixed.

The L2 choke contains 20 turns of a mounting wire with a cross section of 1.5 mm 2, wound on a K28x16x9 ferrite ring.

Transformer T2 is wound on Sh5x5 ferrite with a permeability of 2000 Nm. The halves of the magnetic core are joined with a gap of 0.1 ... 0.2 mm. Winding I contains 180 turns of wire PEV-1 with a diameter of 0.2 mm. Winding II is wound in one row, contains 47 turns of the same wire. Winding III, IV and V each contain 33 turns of wire PEV-1 with a diameter of 0.25 mm. Between the windings you need to lay 2 layers of insulation (for example, masking tape). The phasing of the winding connection is indicated on fig. 2.

Resonant capacitors C1-C10 can only be used with high-quality film capacitors for a voltage of at least 1000 V. It is preferable to use capacitors of the K78-2 type. The blocking capacitor C15 must be of the same type.

The power supply does not need to be adjusted and, with serviceable parts, starts working immediately. It is only necessary to check the voltage values \u200b\u200bfor supplying the drivers 16 ... 17 V. When checking the power supply unit, 220 V mains voltage can be applied to its input terminals GND and +300 V. In the same way, power the power supply unit when setting the resonant frequency.

During the operation of the inverter, all its power elements become hot. The time of continuous operation of the device and its durability will depend on how competently these elements are blown. Heatsinks with a large area must be provided for the VD14-VD21 input rectifier, VT1-VT4 transistors and the VD6, VD7 output rectifier. Forced air cooling is also necessary for the resonant choke L1, the welding transformer T1 and the diodes of the VD10-VD13 doubler. Safety thermostats TS1 and TS2 of the KSD250V type are installed on the radiators of the upper keys and output diodes. All other elements of the inverter do not need blowing and radiators.

Tuning the resonant frequency

To configure the inverter, you need a LATR and a load rheostat with a resistance of 0.15 Ohm. The rheostat must withstand a short-term current flow up to 200 A. The gap of the magnetic circuit of the resonant choke is set to about 1 mm. A jumper is installed between contacts 3 and 4 of the DA8 optocoupler. Install the "wired" microcontroller into the control unit.

The power supply unit must be powered separately during setup. To do this, without connecting the device to the network, the mains voltage of 220 V must be applied to the wires GND and +300 V of the power supply

The power section is still de-energized. After turning on the power, the buzzer should turn on after 5 seconds, then the sound should stop and the relay should turn on. Press both buttons SB1 and SB2 simultaneously. We hold the buttons until the buzzer beeps. Let's release the buttons. The continuous beep will stop and the buzzer will begin to emit an intermittent beep with a period of approximately 2 s. This corresponds to the resonant frequency tuning mode.

If everything is so, then with the help of an oscilloscope we control the presence of bipolar pulses between the gates of transistors VT2 and VT4 with a frequency of 30 kHz with an amplitude of at least 15 V and a step of "dead time" of 2 μs. The same signal should be between the gates VT1 and VT3. If everything is so, we power the power section through the LATR and set the voltage to 20 ... 30 V.

You can turn on a 12 V light bulb to the welding wires. If the light is on, turn on a 0.15 Ohm rheostat and a DC voltmeter in the welding wires. We set the voltage on LATR to 30 ... 40 V and start tuning. Using the SB1 and SB2 buttons, decrease or increase the frequency of the inverter. The limits of frequency change are 30 ... 42 kHz. By adjusting the frequency with the buttons, we achieve the maximum voltage on the rheostat. If the voltage continues to increase with decreasing frequency to 30 kHz, then it is necessary to increase the gap in the magnetic circuit of the resonant choke and repeat the setting again. If, with increasing frequency to 42 kHz, the voltage on the rheostat continues to rise, it is necessary to reduce the gap in the magnetic circuit of the resonant choke and repeat the setting again.

It is necessary to achieve resonance, i.e. adjust the circuit so that increasing or decreasing the frequency of the inverter would lead to a decrease in the voltage across the rheostat. With the elements indicated in the diagram, it is preferable to achieve such a gap in the resonant choke so that resonance with a load of 0.15 Ohm occurs at a frequency of 33 ... 37 kHz. Resonance at a higher frequency will increase the maximum welding current, but the switches and output diodes will operate at their limit.

Once the resonant frequency is tuned, press both buttons simultaneously. After a long sound signal, the value of the resonance frequency will be written to the non-volatile memory of the microcontroller. Turning potentiometer R18, we check the operation of frequency regulation. The minimum frequency must be equal to the resonant frequency. When the potentiometer is turned, the change in frequency should be accompanied by a short beep (17 steps in total).

If everything happens this way, we assemble the entire inverter circuit. Remove the jumper between pins 3 and 4 of the DA8 optocoupler. We connect the inverter to the network. After 5 seconds, the buzzer will sound, then the relay will turn on and the sound will stop. Potentiometer R18 sets the minimum frequency (aka f MAX. POWER), corresponding to the maximum current. We briefly load the inverter with a 0.15 Ohm rheostat and measure the voltage across the load. If this voltage exceeds 23 V, then the setting can be considered complete. If less, then you should increase the gap in the magnetic circuit of the resonant choke and repeat the adjustment from the beginning.